Image processing method, device, system, and storage medium

Abstract

An embodiment of the present disclosure provides an image processing method, comprising: acquiring a real-time video image; selecting a target area on an image frame at a selected moment of the real-time video image; inputting a first image of the target area into an image processing model to obtain a target image; the resolution of the target image is higher than the resolution of the first image; and providing a first interface, displaying the target image on the first interface.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, and in particular to an image processing method, device, system, and storage medium. Background Art

[0002] Security surveillance videos need to achieve ultra-high-definition magnified display of any local area so that local details in the video can be viewed more clearly. Summary of the Invention

[0003] In one aspect, an embodiment of the present disclosure provides an image processing method, comprising:

[0004] Acquire real-time video images;

[0005] Selecting a target area on an image frame at a selected moment of the real-time video image;

[0006] Inputting the first image of the target area into an image processing model to obtain a target image; the resolution of the target image is higher than the resolution of the first image;

[0007] A first interface is provided, and the target image is displayed on the first interface.

[0008] In some embodiments, selecting a target area on an image frame at a selected moment of the real-time video image includes:

[0009] receiving a first instruction input by a user;

[0010] In response to the first instruction, a target area on the image frame at a selected moment of the real-time video image is selected.

[0011] In some embodiments, inputting the first image of the target area into an image processing model to obtain a target image includes:

[0012] Acquire, in time order, a current frame of the real-time video image corresponding to the selected moment of the target area, as well as M frames before and N frames after the current frame; wherein M>N>0, and M and N are both integers;

[0013] Determining a still image and a moving image in the first image based on the current frame and the M frames preceding and N frames following the current frame; wherein the still image is an image of an object in the first image that remains unchanged in position relative to the M frames preceding and N frames following; and the moving image is an image of an object in the first image that changes in position relative to the M frames preceding and N frames following;

[0014] Inputting the first image for determining the still image and the moving image into a cutout algorithm model to obtain independent still images and independent moving images separated from each other;

[0015] Inputting the independent static image and the independent moving image into a super-resolution algorithm model respectively to obtain a target static image and a target moving image;

[0016] The target static image and the target dynamic image are superimposed to obtain the target image.

[0017] In some embodiments, determining the still image and the moving image in the first image based on the current frame and the previous M frames and the next N frames includes:

[0018] Comparing the first image of the current frame with M frames before and N frames after the current frame, determining an image of an object whose position remains unchanged relative to the M frames before and N frames after the current frame in the first image, and determining the image as the still image;

[0019] The first image of the current frame is compared with the M frames before and N frames after the current frame, and the image of the object whose position has changed relative to the M frames before and N frames after the current frame is determined in the first image, and the image is determined as the animated image.

[0020] In some embodiments, determining the still image and the moving image in the first image based on the current frame and the previous M frames and the next N frames includes:

[0021] Taking the time axis as the order, extract a frame from the current frame, the M frames before it, and the N frames after it at set time intervals to obtain X frames; where 0 < X ​​< M + N + 1, and X is an integer;

[0022] Comparing the first image of the current frame with the X frames, determining an image of an object whose position remains unchanged relative to the X frames in the first image, and determining the image as the still image;

[0023] The first image of the current frame is compared with the X frame, and an image of an object whose position has changed relative to the X frame in the first image is determined, and the image is determined as the animated image.

[0024] In some embodiments, determining the still image and the moving image in the first image based on the current frame and the previous M frames and the next N frames includes:

[0025] receiving a still image in the first image input by a user;

[0026] The first image of the current frame and the M frames before and N frames after the current frame are compared with the still picture respectively, and the images of objects whose positions have changed relative to the still picture in the first image and the M frames before and N frames after the current frame are determined as the animated image.

[0027] In some embodiments, determining the still image and the moving image in the first image based on the current frame and the previous M frames and the next N frames includes:

[0028] receiving a still image in the first image input by a user;

[0029] Taking the time axis as the order, extract a frame from the current frame, the M frames before it, and the N frames after it at set time intervals to obtain X frames; where 0 < X ​​< M + N + 1, and X is an integer;

[0030] The first image of the current frame and the X-frame are respectively compared with the still image, and images of objects whose positions have changed relative to the still image in the first image and the X-frame are determined as the animated image.

[0031] In some embodiments, the method further includes: inputting the target image into a data zoom model to obtain an enlarged target image.

[0032] In some embodiments, super-resolution processing is performed on the independent still images at set time intervals in order of the time axis;

[0033] Super-resolution processing is performed on the independent moving image in real time.

[0034] In some embodiments, the resolution of the first image is any one of 2k, 4k and 8k.

[0035] In some embodiments, the target area is a local area or an entire area of ​​an image frame at a selected moment of the real-time video image.

[0036] On the other hand, the present disclosure further provides an image processing method, including:

[0037] Acquire real-time video images;

[0038] Selecting a target area on an image frame at a selected moment of the real-time video image;

[0039] Inputting the first image of the target area into an image processing model to obtain a target image; the resolution of the target image is higher than the resolution of the first image;

[0040] Providing a first interface, displaying the target image on the first interface;

[0041] Inputting the first image of the target area into an image processing model to obtain a target image includes:

[0042] Acquire, in time order, a current frame of the real-time video image corresponding to the selected moment of the target area, as well as M frames before and N frames after the current frame; wherein M>N>0, and M and N are both integers;

[0043] The first image is divided into a first area and a second area; the images in the first area are still images; and the images in the second area are still images and moving images; wherein the still images are images of objects in the first image that remain unchanged in position relative to the first M frames and the next N frames; and the moving images are images of objects in the first image that change in position relative to the first M frames and the next N frames;

[0044] Determine still images and moving images in the second area according to the current frame and the previous M frames and the next N frames;

[0045] Inputting the first image for determining the still image and the moving image into a cutout algorithm model to obtain independent still images and independent moving images separated from each other;

[0046] Inputting the independent static image and the independent moving image into a super-resolution algorithm model respectively to obtain a target static image and a target moving image;

[0047] Superimposing the target static image and the target dynamic image to obtain the target image;

[0048] The determining of still images and moving images in the second area according to the current frame and the previous M frames and the next N frames thereof includes:

[0049] Comparing the image in the second area with the M frames before and N frames after the current frame, determining an image of an object whose position remains unchanged relative to the M frames before and N frames after the current frame in the image in the second area, and determining the image as the still image;

[0050] Comparing the image in the second area with the M frames before and N frames after the current frame, determining an image of an object whose position has changed relative to the M frames before and N frames after the current frame, and determining the image as the animated image;

[0051] Alternatively, a frame is extracted from the current frame, the M frames before it, and the N frames after it at set intervals, in order of the time axis, to obtain X frames; wherein 0<X<M+N+1, and X is an integer;

[0052] Comparing the image in the second area with the X frames, determining an image of an object whose position remains unchanged relative to the X frames in the image in the second area, and determining the image as the still image;

[0053] Comparing the image in the second area with the X frames, determining an image of an object whose position has changed relative to the X frames in the image in the second area, and determining the image as the animated image;

[0054] Alternatively, receiving a still image in the first image input by a user;

[0055] comparing the image in the second area and the M frames before and N frames after the current frame with the still image, and determining as the moving image an image of an object whose position has changed relative to the still image in the image in the second area and the M frames before and N frames after the current frame;

[0056] Alternatively, receiving a still image in the first image input by a user;

[0057] Taking the time axis as the order, extract a frame from the current frame, the M frames before it, and the N frames after it at set time intervals to obtain X frames; where 0 < X ​​< M + N + 1, and X is an integer;

[0058] The image in the second area and the X frames are respectively compared with the still picture, and the image in the second area and the image of the object whose position has changed relative to the still picture in the X frames are determined as the animated image.

[0059] In another aspect, the present disclosure further provides an image processing apparatus, comprising:

[0060] a memory; wherein one or more computer programs are stored in the memory;

[0061] A processor; the processor is coupled to the memory; the processor is configured to execute the computer program to implement the above-mentioned image processing method.

[0062] On the other hand, an embodiment of the present disclosure further provides a non-transitory computer-readable storage medium storing a computer program, wherein when the computer program is executed on a computer, the computer implements the above-mentioned image processing method.

[0063] In another aspect, an embodiment of the present disclosure further provides an electronic device, comprising the above-mentioned image processing device and display device;

[0064] The display device is configured as a first interface.

[0065] In yet another aspect, an embodiment of the present disclosure further provides an image processing system, comprising the above-mentioned image processing apparatus;

[0066] It also includes: a video image acquisition device, a video image transmission processing device and a display device;

[0067] The video image acquisition device acquires real-time video images and transmits them to the video image transmission and processing device;

[0068] The video image transmission processing device receives the real-time video image and transmits it to the image processing device;

[0069] The image processing device processes the real-time video image to obtain a target image, and transmits the target image to the display device;

[0070] The display device receives the target image and displays it.

[0071] In some embodiments, the display device is a projection screen or a display terminal having a resolution consistent with that of the target image. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. The above and other features and advantages will become more apparent to those skilled in the art by describing the detailed exemplary embodiments with reference to the accompanying drawings, in which:

[0073] Figure 1 A flowchart of an image processing method provided in an embodiment of the present disclosure.

[0074] Figure 2 A schematic diagram of an image processing method provided in an embodiment of the present disclosure.

[0075] Figure 3A A schematic diagram of selecting a target area on an image frame of a real-time video image at a selected moment provided by an embodiment of the present disclosure.

[0076] Figure 3B Another schematic diagram of selecting a target area on an image frame of a real-time video image at a selected moment provided by an embodiment of the present disclosure.

[0077] Figure 3C A schematic diagram of another embodiment of the present disclosure for selecting a target area on an image frame of a real-time video image at a selected moment.

[0078] Figure 4This is a flowchart for obtaining a target image in an embodiment of the present disclosure.

[0079] Figure 5 A schematic diagram of a method for determining still images and moving images in a first image provided by an embodiment of the present disclosure.

[0080] Figure 6 A schematic diagram of another method for determining still images and moving images in a first image provided by an embodiment of the present disclosure.

[0081] Figure 7 A schematic diagram of another method for determining still images and moving images in a first image provided by an embodiment of the present disclosure.

[0082] Figure 8 A schematic diagram of another method for determining still images and moving images in a first image provided by an embodiment of the present disclosure.

[0083] Figure 9 A schematic diagram of partitioning a first image provided in an embodiment of the present disclosure.

[0084] Figure 10A A schematic diagram of performing cutout processing on a still image in a first image provided by an embodiment of the present disclosure.

[0085] Figure 10B A schematic diagram of performing cutout processing on a moving image in a first image provided by an embodiment of the present disclosure.

[0086] Figure 11 Schematic diagram of the principles of establishing and applying super-resolution models.

[0087] Figure 12 Schematic diagram of obtaining a target still image and a target moving image after super-resolution processing of an independent still image and an independent moving image.

[0088] Figure 13 A schematic diagram of obtaining a target image by superimposing a target static image and a target dynamic image.

[0089] Figure 14 This is a principle block diagram of an image processing device provided by an embodiment of the present disclosure.

[0090] Figure 15 A block diagram of the principles of an image processing system provided in an embodiment of the present disclosure.

[0091] Figure 16 A topological diagram of an image processing system provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0092] To enable those skilled in the art to better understand the technical solutions of the embodiments of the present disclosure, the image processing method, device, system, and storage medium provided by the embodiments of the present disclosure are further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0093] The embodiments of the present disclosure will be described more fully below with reference to the accompanying drawings, but the illustrated embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully enable those skilled in the art to understand the scope of this disclosure.

[0094] The embodiments of the present disclosure are not limited to the embodiments shown in the drawings, but include modifications of the configurations formed based on the manufacturing process. Therefore, the regions illustrated in the drawings are schematic in nature, and the shapes of the regions shown in the drawings illustrate specific shapes of the regions, but are not intended to be limiting.

[0095] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, "plurality" means two or more.

[0096] When describing some embodiments, the expressions "coupled" and "connected" and their derivatives may be used. For example, when describing some embodiments, the term "connected" may be used to indicate that two or more components are in direct physical or electrical contact with each other. For another example, when describing some embodiments, the term "coupled" may be used to indicate that two or more components are in direct physical or electrical contact. However, the term "coupled" or "communicatively coupled" may also refer to two or more components that are not in direct contact with each other, but still cooperate or interact with each other. The embodiments disclosed herein are not necessarily limited to the contents of this document.

[0097] The use of "adapted to" or "configured to" herein is intended to be open and inclusive language that does not exclude devices adapted or configured to perform additional tasks or steps.

[0098] In related technologies, in security scenarios, video images need to magnify local details to achieve clearer display of local details. Traditional digital zoom will lose clarity while magnifying the image.

[0099] The present disclosure provides an image processing method, referring to Figure 1 and Figure 2 , Figure 1A flowchart of an image processing method provided by an embodiment of the present disclosure; Figure 2 This is a schematic diagram of an image processing method provided by an embodiment of the present disclosure. The image processing method includes:

[0100] Step S1: Acquire real-time video images.

[0101] For example, an 8K real-time video is captured by an ultra-high-definition security camera with a resolution of 8K (7680×4320). Optionally, the real-time video image can also be a 2K (1920×1080) or 4K (4096×2160) video.

[0102] For example, the video image may be a security monitoring video, such as a security monitoring video of a specific location area (an alley corner, a road segment, a specific building at different angles, etc.). Figure 2 As shown in (A) in .

[0103] Step S2: Selecting a target area on an image frame at a selected moment of a real-time video image.

[0104] For example, the selected moment is the moment when the user selects the area of ​​interest. The target area is the area whose details need to be displayed more clearly. For example, the target area is the area where a suspicious situation occurs in the security monitoring video. Figure 2 As shown in (B) in the figure.

[0105] In some embodiments, the target area is a local area or the entire area of ​​the image frame at the selected moment of the real-time video image. Exemplarily, for 2K and 4K video images, the target area can be a local area (such as 1 / 4, 1 / 2, or 1 / 8 area) or the entire area of ​​the image frame at the selected moment. Exemplarily, for 8K video images, the target area is a local area smaller than the entire area of ​​the image frame at the selected moment, for example: 1 / 4, 1 / 2, or 1 / 8 area of ​​the image frame at the selected moment, which is not specifically limited here.

[0106] In some embodiments, selecting a target area on an image frame of a selected moment of a real-time video image includes: providing a second interface, the second interface being used to display an indication of user input, such as a first instruction input by the user; receiving the first instruction input by the user; and selecting the target area on the image frame of the selected moment of the real-time video image in response to the first instruction.

[0107] The selected time is determined according to the first instruction input by the user. Exemplarily, the selected time may be the time when the first instruction input by the user is received; the selected time may also be a time point defined by the user.

[0108] Exemplarily, the second interface is an image frame interface at a selected moment; the second interface and the first interface displaying the target image can be two different display interfaces or the same display interface. Exemplarily, the first command input by the user includes: a gesture command, a voice command, or an infrared remote control command.

[0109] For example, refer to Figure 3A , which is a schematic diagram of selecting a target area on an image frame at a selected moment of a real-time video image; in the case where the first instruction is a voice instruction, the second interface 201 may display a first prompt 2011: "Please say 'selection start'", and the user inputs the voice instruction "selection start", referring to Figure 3A , a selection box icon 2012 is displayed on the second interface 201, and the user can set the position and size of the framed area of ​​the selection box icon 2012 to select a target area of ​​interest on the video image.

[0110] For example, refer to Figure 3B , which is another schematic diagram for selecting a target area on an image frame at a selected moment in a real-time video image. When the first instruction is a gesture instruction, a first prompt 2011 may be displayed on the second interface 201: "Use 'selection box' gesture to start selection." The user enters the "selection box" gesture, and a selection box icon 2012 appears on the second interface 201. The user can set the position and size of the area framed by selection box icon 2012, thereby selecting the target area of ​​interest on the video image. Optionally, the gesture instruction can also be other common gestures such as "scissor hand" or "thumbs up," and the first prompt displayed on the second interface can be adjusted accordingly. This is not specifically limited here.

[0111] Optional, see Figure 3C , which is another schematic diagram of selecting a target area on an image frame at a selected moment in a real-time video image. The user can also trigger the selection by pressing a button. For example, the user presses the "OK" button on the remote control, and a selection box button 2013 appears on the second interface 201. The user clicks the selection box button 2013, and a selection box icon 2012 appears on the second interface 201. The user can set the position and size of the area defined by the selection box icon 2012, thereby selecting the target area of ​​interest on the video image. Optionally, the button on the remote control for determining the selection can also be other buttons such as the "Selection" button and the "Confirm" button, which are not specifically limited here.

[0112] In some other embodiments, the above-mentioned buttons can be physical buttons or virtual buttons on a touch screen, and the remote control can be an infrared emitting device with physical buttons or an electronic device with a touch screen and infrared emitting function, which is not limited here.

[0113] It should be noted that the voice command, gesture command or key command corresponding to the first command can be set in advance or customized by the user. After the user enters the first command, whether to display the checkbox icon on the second interface 201 can also be customized by the user, or the checkbox icon may not be displayed.

[0114] Step S3: inputting the first image of the target area into the image processing model to obtain a target image; the resolution of the target image is higher than the resolution of the first image.

[0115] The first image is the entire image within the target area selected from the image frame at the selected moment of the real-time video image.

[0116] In some embodiments, reference Figure 4 , which is a flow chart for obtaining a target image in an embodiment of the present disclosure. A first image of a target area is input into an image processing model to obtain a target image; the flow chart includes:

[0117] Step S31: obtaining the current frame of the real-time video image corresponding to the selected moment of the target area and the M frames before and N frames after the current frame in order of time axis; wherein M>N>0, and M and N are both integers.

[0118] The current frame of the real-time video image corresponding to the moment of selecting the target area is the image frame of the real-time video image at the selected moment, and the first image is the first image of the target area on the current frame.

[0119] For example, frames within 10 seconds before and 2 seconds after the current frame are obtained, and 25-30 frames are refreshed per second. Frames before and after the current frame can also be selected from frames within 20 seconds or 10 minutes before and 5 seconds or 2 minutes after the current frame. The selection of frames before and after the current frame is not limited here.

[0120] It should be noted that, the more frames are selected, the more accurate the determination of the static image and the dynamic image in the first image is, and thus the better the effect of the target image obtained subsequently is.

[0121] Step S32: Determine the still image and the moving image in the first image based on the current frame and the previous M frames and the next N frames; wherein the still image is an image of an object in the first image whose position remains unchanged relative to the previous M frames and the next N frames; and the moving image is an image of an object in the first image whose position changes relative to the previous M frames and the next N frames.

[0122] In some embodiments, step S32 includes: step S321: comparing a first image of the current frame with M frames before and N frames after the current frame, determining an image of an object whose position remains unchanged relative to the M frames before and N frames after the current frame in the first image, and determining the image as a still image;

[0123] Step S322: Compare the first image of the current frame with the M frames before and N frames after the current frame, determine the image of the object whose position has changed relative to the M frames before and N frames after the current frame, and determine it as a moving image.

[0124] For example, Figure 5 FIG2 is a schematic diagram of a method for determining whether a static image or a moving image is present in a first image. In the current frame, the M frames preceding it, and the N frames following it, the position of the "Forbidden City Museum Tower" in first image 300 remains unchanged relative to the M frames preceding it and the N frames following it; therefore, the "Forbidden City Museum Tower" in first image 300 is determined as a static image 301. However, the position of the "balloon" in first image 300 changes relative to the M frames preceding it and the N frames following it; therefore, the "balloon" in first image 300 is determined as a moving image 302.

[0125] In steps S321-S322, the method for determining still and moving images involves comparing the first image of the current frame with the M frames preceding and N frames following the current frame. If an object in the first image changes position relative to at least two frames, the image of that object is determined to be a moving image. Objects in the first image whose position remains constant relative to all selected frames are determined to be still images. While this method for determining still and moving images is relatively complex and time-consuming, it provides a relatively accurate distinction between still and moving images.

[0126] In some embodiments, step S32 includes: step S321': extracting a frame from the current frame and its previous M frames and subsequent N frames at set time intervals in order of the time axis to obtain X frames; where 0<X<M+N+1, and X is an integer.

[0127] Step S322 ′: Compare the first image of the current frame with the X frame, determine the image of the object whose position in the first image remains unchanged relative to the X frame, and determine it as a still image.

[0128] Step S323 ′: Compare the first image of the current frame with the X frame, determine the image of the object whose position in the first image has changed relative to the X frame, and determine it as a moving image.

[0129] For example, Figure 6FIG2 is a schematic diagram of another method for determining whether a still image or a moving image is present in a first image. A frame is extracted from the current frame, the M frames preceding it, and the N frames following it, in chronological order, at one-second intervals, for a total of X frames (wherein X frames may be extracted from a longer period of 10-20 minutes or more than 30 minutes). A first image 300 of the current frame is compared with the X frames. The position of the "Forbidden City Museum Tower" in the first image 300 relative to the X frames remains unchanged. Therefore, the "Forbidden City Museum Tower" in the first image 300 is determined as a still image 301. The position of the "balloon" in the first image 300 relative to the X frames changes. Therefore, the "balloon" in the first image 300 is determined as a moving image 302.

[0130] In steps S321'-S323', the method for determining still and moving images involves extracting X frames from the current frame, the M frames preceding it, and the N frames following it at equal time intervals. The first image of the current frame is then compared against these X frames. If an object in the first image changes position relative to images in at least two frames, the image of that object is determined to be a moving image. Objects in the first image whose position remains unchanged relative to the extracted X frames are determined to be still images. This method for determining still and moving images not only simplifies the comparison process and reduces time consumption, but also produces relatively accurate results for determining still and moving images.

[0131] In some embodiments, step S32 includes: S321": receiving a still image in the first image input by a user.

[0132] S322: Compare the first image of the current frame and the M frames before and N frames after the current frame with the still image, and determine as a moving image the images of objects whose positions have changed relative to the still image in the first image and the M frames before and N frames after the current frame.

[0133] For example, Figure 7 FIG2 is a schematic diagram of another method for determining still images and moving images in a first image. A still image 303 in a first image, such as "Forbidden City Museum Tower," is received as input by a user. The first image 300 of the current frame, as well as the M frames preceding and N frames following the current frame, are then compared one by one with the still image 303 in the order of frame refresh time. Images of objects in the first image 300 and the M frames preceding and N frames following the current frame whose positions relative to the still image 303 change are determined as moving images 302. Accordingly, images of objects in the first image 300 and the M frames preceding and N frames following the current frame whose positions relative to the still image 303 remain unchanged are determined as still images 301.

[0134] In steps S321-S322, the method for determining still and moving images is to input a still image in the first image as a comparison object. The first image of the current frame, as well as the M frames preceding and N frames following the current frame, are then compared with the still image one by one. If an object in the first image changes position relative to the still image, the image of that object is determined to be a moving image. Images of objects in the first image whose position remains unchanged relative to the still image are determined to be still images. This method for determining still and moving images also simplifies the comparison process and takes less time, while also obtaining relatively accurate still and moving image determination results.

[0135] In some embodiments, step S32 includes: S321 ′′′: receiving a still image in the first image input by a user.

[0136] S322′′′: extracting a frame from the current frame, the previous M frames, and the next N frames at set intervals in order of the time axis to obtain X frames; wherein 0<X<M+N+1, and X is an integer.

[0137] S323′′′: Compare the first image of the current frame and the X-frame image with the still image, and determine as a moving image any image whose position in the first image and the X-frame image relative to the still image has changed. Accordingly, determine as a still image any image whose position in the first image and the X-frame image relative to the still image has remained unchanged.

[0138] For example, Figure 8 FIG. 1 is a schematic diagram of another method for determining whether a static image or a moving image in a first image. A user input of a still image 303 in a first image, such as "Forbidden City Museum Tower," is received. A frame is extracted from a current frame, its preceding M frames, and its succeeding N frames at 1-second intervals, in chronological order, for a total of X frames (wherein X frames may be extracted from a number of frames within a longer period of 10-20 minutes or more than 30 minutes). A first image 300 of the current frame and X frames are compared one by one with the still image 303. The position of the "Forbidden City Museum Tower" in the first image 300 and X frames remains unchanged relative to the still image 303. Therefore, the "Forbidden City Museum Tower" in the first image 300 is determined as a static image 301. A "balloon" in the first image 300 and X frames changes position during one or more comparisons with the still image 303. Therefore, the "balloon" in the first image 300 is determined as a moving image 302.

[0139] Among them, the solution for determining the static image and the dynamic image in steps S321'''-S323''' is to Figure 6 and Figure 7 The scheme for determining static images and dynamic images is combined with the scheme for determining static images and dynamic images in the method, which can further simplify the comparison process, shorten the comparison time, and further make the determination results of static images and dynamic images more accurate.

[0140] In some embodiments, step S32 includes: Figure 9 The figure shows a schematic diagram of partitioning a first image. The first image 300 is divided into a first area 304 and a second area 305; the image in the first area 304 is a still image 301; the image in the second area 305 is a still image 301 and a moving image 302; the still image 301 and the moving image 302 in the image in the second area 305 are determined based on the current frame and the previous M frames and the next N frames; specifically, the image in the second area 305 is executed Figure 5-Figure 8 Any of the above processing methods can be used to determine the static image 301 and the moving image 302 in the second area 305. This can further simplify the comparison process, shorten the comparison time, and further make the determination result of the static image and the moving image more accurate.

[0141] In some embodiments, the image in the second zone 305 is executed Figure 5 The processing method in the embodiment of the present invention is used to determine the static image 301 and the moving image 302 in the second area 305, including: comparing the image in the second area 305 with the M frames before and the N frames after the current frame, determining the image of an object whose position remains unchanged relative to the M frames before and the N frames after the current frame, and determining the image as the static image 301; comparing the image in the second area 305 with the M frames before and the N frames after the current frame, determining the image of an object whose position changes relative to the M frames before and the N frames after the current frame, and determining the image as the moving image 302.

[0142] In some embodiments, the image in the second zone 305 is executed Figure 6 The processing method for determining the still image 301 and the moving image 302 in the second area 305 includes: extracting a frame from the current frame, the M frames before it, and the N frames after it at set time intervals in order of the time axis to obtain X frames, where 0<X<M+N+1, and X is an integer; comparing the image in the second area 305 with the X frames, determining an image of an object whose position remains unchanged relative to the X frames in the images in the second area 305, and determining the image as the still image 301; comparing the image in the second area 305 with the X frames, determining an image of an object whose position changes relative to the X frames in the images in the second area 305, and determining the image as the moving image 302.

[0143] In some embodiments, the image in the second zone 305 is executed Figure 7The processing method in the embodiment of the present invention is to determine the still image 301 and the moving image 302 in the second area 305, including: receiving the still image 303 in the first image 300 input by the user; comparing the image in the second area 305 and the M frames before and N frames after the current frame with the still image 303 respectively, and determining the image in the second area 305 and the M frames before and N frames after the current frame whose positions relative to the still image 303 have changed as the moving image 302; correspondingly, determining the image in the second area 305 and the M frames before and N frames after the current frame whose positions relative to the still image 303 remain unchanged as the still image 301.

[0144] In some embodiments, the image in the second zone 305 is executed Figure 8 The processing method for determining the still image 301 and the moving image 302 in the second area 305 includes: receiving a still image 303 in the first image 300 input by a user; extracting a frame from the current frame, the M frames before it, and the N frames after it at set time intervals in order of the time axis to obtain X frames, where 0<X<M+N+1, and X is an integer; comparing the image in the second area 305 and the X frames with the still image 303, and determining the image in the second area 305 and the image of the object in the X frames whose position relative to the still image 303 changes as the moving image 302; correspondingly, determining the image in the second area 305 and the image of the object in the X frames whose position relative to the still image 303 remains unchanged as the still image 301.

[0145] Step S33: inputting the first image for determining the still image and the moving image into a cutout algorithm model to obtain independent still images and independent moving images separated from each other.

[0146] For example, referring to Figure 10A , is a schematic diagram of performing a cutout process on a still image in a first image. The independent still image may be a mask image of the still image in the first image (refer to Figure 10A As shown in (B) in the figure). For example, the value range of each pixel in the independent still image is [0,1], which reflects the probability that the pixel point of the first image is a still image. It can be understood that in the center area of ​​the still image of the first image, the value of the independent still image is 1, and in the area outside the still image of the first image, the value of the independent still image is 0. In the boundary area between the still image and the area outside the still image, the value of the independent still image is between 0 and 1, indicating that the pixel point in the boundary area may be a still image, a moving image, or both. For example, the independent still image can refer to Figure 10A As shown in (D),

[0147] For example, referring to Figure 10B, is a schematic diagram of performing a cutout process on the animated image in the first image. The independent animated image can be a mask image of the animated image in the first image (refer to Figure 10B As shown in (C) in the figure). For example, the value range of each pixel in the independent animation is [0,1], which reflects the probability that the pixel point of the first image is an animation. It can be understood that in the center area of ​​the animation of the first image, the value of the independent animation is 1, and in the area outside the animation of the first image, the value of the independent animation is 0. In the boundary area between the animation and the area outside the animation, the value of the independent animation is between 0 and 1, indicating that the pixel point in the boundary area may be an animation, a static image, or both. For example, the independent animation can refer to Figure 10B As shown in (E) in .

[0148] Exemplarily, the cutout model is a pre-trained neural network model. For example, the cutout model may be a salient object detection (SOD) model. The salient object detection model may be a U2-Net model. The salient object detection model is used to distinguish the most attractive targets in an image. During the training process, the salient object detection model is trained using a specific type of training set, so that the target object segmentation model for a specific scene can be trained in a short time. For example, a large number of static images are used to train the salient object detection model, and the resulting cutout model is more efficient in the segmentation processing of static images, and the segmentation effect is also better. For another example, a large number of moving images are used to train the salient object detection model, and the resulting cutout model is more efficient in the segmentation processing of moving images, and the segmentation effect is also better.

[0149] In some other embodiments, when the target object is a still image or a moving image, the image processing model may also be other neural network models for still image or moving image segmentation, such as a deep convolutional neural network (DCNN) model for portrait segmentation reasoning. The DCNN model can perform portrait mask reasoning on the first image, and the mask image output by the DCNN model is a portrait moving image.

[0150] In some embodiments, the first image is input into a cutout algorithm model that matches the type of the first image. For example, if the first image is a still image, then the cutout algorithm model's training set should be still images, as the model's segmentation effect for still images is relatively accurate. For another example, if the first image is a moving image, then the cutout algorithm model's training set should be moving images, as the model's segmentation effect for moving images is relatively accurate.

[0151] In some embodiments, based on the resolution of the first image, the first image is input into a cutout algorithm model that matches the resolution of the first image. For example, if the resolution of the first image is very high, a cutout algorithm model with a higher cutout resolution is selected. For another example, if the resolution of the first image is significantly lower than the resolution of the image required as input by the cutout algorithm model, if the interpolation is greater than 100% of the size of the first image, excessive interpolation may result in image distortion, and therefore a cutout algorithm model with a lower cutout resolution may be selected.

[0152] In some embodiments, the matting algorithm model may also adopt any one of a Trimap-based model, a Deep ImageMatting model, a Background Matting model, a Background Matting V2 model, a Trimap-free model, a Semantic Human Matting model, and a Modnet model.

[0153] Step S34: inputting the independent still image and the independent moving image into the super-resolution algorithm model respectively to obtain the target still image and the target moving image.

[0154] For example, super-resolution is a low-resolution image processing task that maps a low-resolution image to a high-resolution image, and enlarges a low-resolution image into a high-resolution image in order to enhance image details. For example, the resolution of the first image is any one of 2k, 4k and 8k. The first image includes an independent still image and an independent moving image. The first image is as follows: Figure 2 As shown in (C), the target image is Figure 2 As shown in (D).

[0155] For example, the super-resolution model can be a deep learning model such as DUF, EDVR, RFDN, UNet, etc. Figure 11 This is a schematic diagram illustrating the principles of establishing and applying a super-resolution algorithm model. A deep learning model first accumulates and learns from a large number of high-resolution images, then applies the learning model to low-resolution images for restoration. Finally, it recovers the high-frequency details of the image, achieving better image restoration and improving image recognition capabilities and accuracy.

[0156] The training principle of the super-resolution algorithm model is as follows: (1) First, the high-resolution image is degraded according to the degradation model to generate a training model.

[0157] (2) Divide the image into blocks according to the correspondence between the low-frequency part and the high-frequency part of the high-resolution image, learn through a certain algorithm, obtain prior knowledge, and establish a learning model.

[0158] (3) Based on the input low-resolution block, search for the best matching high-frequency block in the established training set.

[0159] In some embodiments, reference Figure 12 , is a schematic diagram of the target static image and target dynamic image obtained after the independent static image and the independent dynamic image are processed by super resolution. Figure 10A (D) in the input super-resolution algorithm model, after super-resolution processing, the target static image is obtained, reference Figure 12 (F) in the target static image; the resolution of the target static image is higher than that of the independent static image; the independent dynamic image ( Figure 10B (E) in the input super-resolution algorithm model, after super-resolution processing, the target dynamic image is obtained, reference Figure 12 (G) in the figure; the resolution of the target animation is higher than that of the independent animation.

[0160] In some embodiments, super-resolution processing is performed on independent static images at set intervals along the timeline; super-resolution processing is performed on independent moving images in real time. For example, super-resolution processing may be performed on independent static images at intervals of 1 second or 10 seconds. Alternatively, super-resolution processing may be performed on independent static images at intervals of 5 seconds or 15 seconds, or real-time super-resolution processing may be performed on independent static images. The intervals set are not specifically limited and can be set by the user at will.

[0161] Step S35: superimpose the target static image and the target dynamic image to obtain a target image.

[0162] For example, the target static image and the target dynamic image are superimposed, referring to Figure 13 , is a schematic diagram of superimposing the target static image and the target dynamic image to obtain the target image. Among them, the target static image is Figure 12 In (F), the target animation is Figure 12 (G) in the image, the target image obtained after superposition is Figure 13 (H) in the , including:

[0163] 1) Image merging.

[0164] According to the mask image, the target static image and the target dynamic image are merged. The merged image satisfies the formula:

[0165] I fusion1 =I person ×I mask +I background ×(lI mask ).

[0166] Among them, I fusion1 represents the merged image, I person Indicates the target animation, I background Represents the target static image, Imask represents the mask image, I mask The value range of each element in is [0, 1].

[0167] It can be understood that the mask image is traversed pixel by pixel, the target moving image area of ​​the first image is intercepted, and collage is made into the target static image. The target moving image covers the original image of the corresponding area of ​​the target static image to obtain a merged image. Since the value of the mask image in the boundary area between the target moving image and the target static image in the first image is between 0 and 1, it has a certain degree of transparency, rather than a clear boundary such as non-zero or one. Therefore, the merged image will transition naturally at the edge. For example, if the value of a certain pixel point at the boundary of the mask image is 0.3, then the merged image is a fusion of the target moving image with a value of 0.3 and the target static image with a value of 0.7 at that pixel point, so the boundary transition is natural and not abrupt.

[0168] 2) Image fusion.

[0169] It is understandable that the brightness, saturation, clarity, etc. of the target moving image and the target still image may be inconsistent. Therefore, the merged image is subjected to one or more fusion processing operations including brightness adjustment, contrast adjustment, color adjustment, saturation adjustment, and Gaussian filtering to generate a fused image of the target moving image and the target still image.

[0170] For example, assuming that the brightness of the target moving image is 50 and the brightness of the target static image is 100, the brightness of the merged image is adjusted to 80, so that the overall brightness of the fused image is consistent, and there is no obvious boundary or brightness difference between the target moving image and the target static image.

[0171] For example, the resolution of the target still image is 7680×4320, and the resolution of the target moving image is also 7680×4320.

[0172] It can be understood that one or more of the above-mentioned fusion processing operations of brightness adjustment, contrast adjustment, color adjustment, saturation adjustment, Gaussian filtering, etc. can be achieved by using the corresponding image processing function to perform overall processing on the merged image, or by using the corresponding image processing function to perform separate processing on the target dynamic image or target static image.

[0173] In some embodiments, the image processing method further includes step S36: inputting the target image into a data zoom model to obtain an enlarged target image.

[0174] Digital zoom, also known as digital zoom, uses the camera's processor to magnify each pixel within an image. This technique is similar to using image processing software to enlarge an image, but the program operates within the camera, magnifying a portion of the original CCD image sensor's pixels using a "interpolation" process. This algorithm then magnifies the entire image.

[0175] Step S4: providing a first interface, and displaying the target image on the first interface.

[0176] For example, refer to Figure 13 As shown in (H) in FIG. 8 , the first interface 202 displays a target image.

[0177] The above-mentioned image processing method can perform cutout, super-resolution, overlay and data zoom processing on local areas of ultra-high-definition video, thereby further obtaining ultra-high-definition video of the local area, allowing users to see smaller details in the local area of ​​the ultra-high-definition video in security scenarios, thereby improving the user experience; at the same time, this image processing method does not require users to perform unnecessary operations, thereby improving image processing efficiency.

[0178] The present disclosure also provides an image processing device, referring to Figure 14 , which is a block diagram of the principles of an image processing device provided in an embodiment of the present disclosure. The image processing device includes: a memory 401 storing one or more computer programs; a processor 402 coupled to the memory 401; and the processor 402 is configured to execute the computer programs to implement the above-described image processing method.

[0179] Exemplarily, the processor 402 and the memory 401 are coupled via, for example, an I / O interface, thereby enabling information interaction.

[0180] Exemplarily, the processor 402 may be a processor or a general term for multiple processing elements. For example, the processor 402 may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the disclosed solution, such as one or more microprocessors. For another example, the processor 402 may be a programmable device; for example, the programmable device may be a CPLD (Complex Programmable Logic Device), an EPLD (Erasable Programmable Logic Device), or an FPGA (field-programmable gate array).

[0181] The memory 401 may be a single memory or a collective term for multiple memory elements, and is used to store executable program codes, etc. The memory 401 may include a random access memory or a non-volatile memory, such as a disk memory or a flash memory.

[0182] The memory 401 is used to store application code for executing the solution of the present disclosure, and is controlled by the processor 402. The processor 402 is used to execute the application code stored in the memory 401 to control the image processing apparatus to implement the image processing method provided by any of the above embodiments of the present disclosure.

[0183] The beneficial effects of the above-mentioned image processing device are the same as the beneficial effects of the image processing methods described in some of the above-mentioned embodiments, and will not be repeated here.

[0184] The embodiments of the present disclosure also provide a non-transitory computer-readable storage medium (e.g., a non-transitory computer-readable storage medium) storing a computer program, wherein when the computer program is executed on a computer, the computer implements the image processing method of any of the above embodiments.

[0185] Exemplarily, the above-mentioned computer-readable storage media may include, but are not limited to: magnetic storage devices (e.g., hard disks, floppy disks, or magnetic tapes, etc.), optical disks (e.g., CDs (Compact Disks), DVDs (Digital Versatile Disks), etc.), smart cards, and flash memory devices (e.g., EPROMs (Erasable Programmable Read-Only Memory), cards, sticks, or key drives, etc.). The various computer-readable storage media described in the present disclosure may represent one or more devices and / or other machine-readable storage media for storing information. The term "machine-readable storage medium" may include, but is not limited to, wireless channels and various other media capable of storing, containing, and / or carrying instructions and / or data.

[0186] The beneficial effects of the above-mentioned non-transitory computer-readable storage medium are the same as the beneficial effects of the image processing methods described in some of the above embodiments, and will not be repeated here.

[0187] An embodiment of the present disclosure further provides an electronic device, comprising the image processing device and the display device in the above embodiment; the display device is configured as a first interface.

[0188] Among them, the display device is a high-resolution (such as 8K) display terminal. For example, the display device can project the target image displayed on the first interface to an outdoor 8K large screen. The display device can also be an 8K TV.

[0189] The beneficial effects of the electronic device are the same as the beneficial effects of the image processing methods described in some of the above embodiments, and will not be repeated here.

[0190] The present disclosure also provides an image processing system, referring to Figure 15 and Figure 16 , Figure 15 A block diagram of an image processing system according to an embodiment of the present disclosure; Figure 16 This is a topology diagram of an image processing system provided by an embodiment of the present disclosure. The system includes the aforementioned image processing device 501; a video image acquisition device 502, a video image transmission and processing device 503, and a display device 504. The video image acquisition device 502 acquires real-time video images and transmits them to the video image transmission and processing device 503. The video image transmission and processing device 503 receives the real-time video images and transmits them to the image processing device 501. The image processing device 501 processes the real-time video images to obtain target images, and transmits the target images to the display device 504. The display device 504 receives the target images and displays them.

[0191] In some embodiments, the display device 504 is a projection screen or display terminal with a resolution consistent with the target image. For example, the video image acquisition device 502 utilizes multiple 8K ultra-high-definition security cameras. The video image transmission and processing device 503 utilizes a content distribution network streaming media server (i.e., an ultra-high-definition video server). The image processing device 501 and the display device 504 utilize a video stream projection control workstation and a projection screen, respectively.

[0192] Exemplarily, the specific workflow of the image processing system is as follows: a camera is used to capture 8K ultra-high-definition security video images, which are encoded into an H.265 (compressed video coding standard) video stream, and then the video stream is pushed to an 8K ultra-high-definition content distribution network streaming server through the RTMP (Real Time Messaging Protocol) or RTSP (Real Time Streaming Protocol, RFC2326, Real-time Streaming Protocol) network protocol; then the video stream projection control workstation obtains the video stream from the 8K ultra-high-definition content distribution network streaming server through HTTP (Hyper Text Transfer Protocol, HTTP, Hypertext Transfer Protocol) or RTMP network protocol, performs image processing on it to obtain the target image, and projects the target image onto the 8K large screen.

[0193] The following is a detailed description of the functions of the four devices in the image processing system:

[0194] 8K UHD security cameras: These cameras capture and encode 8K video, outputting video streams in 2K (1920×1080), 4K (4096×2160), or 8K (7680×4320) resolutions using the H.265 encoding format. These streams are then pushed to an UHD content delivery network streaming server over the network. In some embodiments, multiple 8K UHD security cameras can be deployed, each capturing 8K video images.

[0195] Content Distribution Network Streaming Server: This server is primarily responsible for receiving and forwarding 2K, 4K, or 8K video streams. Features include: efficient H.265 format, 8K ultra-high bitrate processing engine, high-concurrency framework design, flexible data transmission, port security management, multiple hotlink protection features, intelligent data management, and data visualization modules.

[0196] Video Stream Projection Control Workstation: This is primarily responsible for decoding, displaying, and outputting 2K, 4K, or 8K video streams. Features include: 8K large-format real-time projection, real-time selection tracking, multi-channel big data rendering, seamless rendering and blending technology, real-time GPU rendering acceleration, high-speed I / O transmission, real-time tagging, layer adjustment and processing, real-time color adjustment, and remote camera control. In some embodiments, multiple video stream projection control workstations may be deployed, each performing image processing and output for a different video stream.

[0197] Projection screen: For example, an 8K large screen, primarily responsible for the terminal display of the 8K video stream, such as an outdoor 8K large screen or 8K TV. In some embodiments, multiple projection screens can be used, each corresponding to the video stream output by a different video stream projection control workstation.

[0198] The beneficial effects of the above-mentioned image processing system are the same as the beneficial effects of the image processing methods described in some of the above-mentioned embodiments, and will not be repeated here.

[0199] It is understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present disclosure, and the present disclosure is not limited thereto. Those skilled in the art may make various modifications and improvements without departing from the spirit and substance of the present disclosure, and such modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. An image processing method, comprising: Acquire real-time video images; Selecting a target area on an image frame at a selected moment of the real-time video image; Inputting the first image of the target area into an image processing model to obtain a target image; the resolution of the target image is higher than the resolution of the first image; Providing a first interface, displaying the target image on the first interface; Inputting the first image of the target area into an image processing model to obtain a target image, comprising: Acquire, in time order, a current frame of the real-time video image corresponding to the selected moment of the target area, as well as M frames before and N frames after the current frame; wherein M>N>0, and M and N are both integers; Determining a still image and a moving image in the first image based on the current frame and the M frames preceding and N frames following the current frame; wherein the still image is an image of an object in the first image that remains unchanged in position relative to the M frames preceding and N frames following; and the moving image is an image of an object in the first image that changes in position relative to the M frames preceding and N frames following; Inputting the first image for determining the still image and the moving image into a cutout algorithm model to obtain independent still images and independent moving images separated from each other; Inputting the independent static image and the independent moving image into a super-resolution algorithm model respectively to obtain a target static image and a target moving image; The target static image and the target dynamic image are superimposed to obtain the target image.

2. The image processing method according to claim 1, wherein: The step of selecting a target area on an image frame at a selected moment of the real-time video image comprises: receiving a first instruction input by a user; In response to the first instruction, a target area on the image frame at a selected moment of the real-time video image is selected.

3. The image processing method according to claim 1, wherein: The determining of the still image and the moving image in the first image according to the current frame and the previous M frames and the next N frames thereof includes: Comparing the first image of the current frame with M frames before and N frames after the current frame, determining an image of an object whose position remains unchanged relative to the M frames before and N frames after the current frame in the first image, and determining the image as the still image; The first image of the current frame is compared with the M frames before and N frames after the current frame, and the image of the object whose position has changed relative to the M frames before and N frames after the current frame is determined in the first image, and the image is determined as the animated image.

4. The image processing method according to claim 1, wherein: The determining of the still image and the moving image in the first image according to the current frame and the previous M frames and the next N frames thereof includes: Taking the time axis as the order, extract a frame from the current frame, the M frames before it, and the N frames after it at set time intervals to obtain X frames; where 0 < X ​​< M + N + 1, and X is an integer; Comparing the first image of the current frame with the X frames, determining an image of an object whose position remains unchanged relative to the X frames in the first image, and determining the image as the still image; The first image of the current frame is compared with the X frame, and an image of an object whose position has changed relative to the X frame in the first image is determined, and the image is determined as the animated image.

5. The image processing method according to claim 1, wherein: The determining of the still image and the moving image in the first image according to the current frame and the previous M frames and the next N frames thereof includes: receiving a still image in the first image input by a user; The first image of the current frame and the M frames before and N frames after the current frame are compared with the still picture respectively, and the images of objects whose positions have changed relative to the still picture in the first image and the M frames before and N frames after the current frame are determined as the animated image. The image processing method according to claim 1 , wherein: The determining of the still image and the moving image in the first image according to the current frame and the previous M frames and the next N frames thereof includes: receiving a still image in the first image input by a user; Taking the time axis as the order, extract a frame from the current frame, the M frames before it, and the N frames after it at set time intervals to obtain X frames; where 0 < X ​​< M + N + 1, and X is an integer; The first image of the current frame and the X-frame are respectively compared with the still image, and images of objects whose positions have changed relative to the still image in the first image and the X-frame are determined as the animated image.

7. The image processing method according to claim 1, wherein: Also includes: The target image is input into a data zoom model to obtain an enlarged target image.

8. The image processing method according to claim 1, wherein: performing super-resolution processing on the independent still images at set time intervals in order of the time axis; Super-resolution processing is performed on the independent moving image in real time.

9. The image processing method according to claim 1, wherein: The resolution of the first image is any one of 2k, 4k and 8k.

10. The image processing method according to claim 1, wherein: The target area is a local area or an entire area of ​​an image frame of the real-time video image at a selected moment.

11. An image processing method, comprising: Acquire real-time video images; Selecting a target area on an image frame at a selected moment of the real-time video image; Inputting the first image of the target area into an image processing model to obtain a target image; the resolution of the target image is higher than the resolution of the first image; Providing a first interface, displaying the target image on the first interface; Inputting the first image of the target area into an image processing model to obtain a target image includes: Acquire, in time order, a current frame of the real-time video image corresponding to the selected moment of the target area, as well as M frames before and N frames after the current frame; wherein M>N>0, and M and N are both integers; The first image is divided into a first area and a second area; the images in the first area are still images; and the images in the second area are still images and moving images; wherein the still images are images of objects in the first image that remain unchanged in position relative to the first M frames and the next N frames; and the moving images are images of objects in the first image that change in position relative to the first M frames and the next N frames; Determine still images and moving images in the second area according to the current frame and the previous M frames and the next N frames; Inputting the first image for determining the still image and the moving image into a cutout algorithm model to obtain independent still images and independent moving images separated from each other; Inputting the independent static image and the independent moving image into a super-resolution algorithm model respectively to obtain a target static image and a target moving image; Superimposing the target static image and the target dynamic image to obtain the target image; The determining of still images and moving images in the second area according to the current frame and the previous M frames and the next N frames thereof includes: Comparing the image in the second area with the M frames before and N frames after the current frame, determining an image of an object whose position remains unchanged relative to the M frames before and N frames after the current frame in the image in the second area, and determining the image as the still image; Comparing the image in the second area with the M frames before and N frames after the current frame, determining an image of an object whose position has changed relative to the M frames before and N frames after the current frame, and determining the image as the animated image; Alternatively, a frame is extracted from the current frame, the M frames before it, and the N frames after it at set intervals, in order of the time axis, to obtain X frames; wherein 0<X<M+N+1, and X is an integer; Comparing the image in the second area with the X frames, determining an image of an object whose position remains unchanged relative to the X frames in the image in the second area, and determining the image as the still image; Comparing the image in the second area with the X frames, determining an image of an object whose position has changed relative to the X frames in the image in the second area, and determining the image as the animated image; Alternatively, receiving a still image in the first image input by a user; comparing the image in the second area and the M frames before and N frames after the current frame with the still image, and determining as the moving image an image of an object whose position has changed relative to the still image in the image in the second area and the M frames before and N frames after the current frame; Alternatively, receiving a still image in the first image input by a user; Taking the time axis as the order, extract a frame from the current frame, the M frames before it, and the N frames after it at set time intervals to obtain X frames; where 0 < X ​​< M + N + 1, and X is an integer; The image in the second area and the X frames are respectively compared with the still picture, and the image in the second area and the image of the object whose position has changed relative to the still picture in the X frames are determined as the animated image.

12. An image processing apparatus, comprising: Memory; described One or more computer programs are stored in the memory; processor; The processor is coupled to the memory; The processor is configured to execute the computer program to implement the image processing method according to any one of claims 1 to 11.

13. A non-transitory computer-readable storage medium storing a computer program, wherein: When the computer program is executed, the computer implements the image processing method according to any one of claims 1 to 11.

14. An image processing system, wherein: comprising the image processing device as claimed in claim 12; It also includes: a video image acquisition device, a video image transmission processing device and a display device; The video image acquisition device acquires real-time video images and transmits them to the video image transmission and processing device; The video image transmission processing device receives the real-time video image and transmits it to the image processing device; The image processing device processes the real-time video image to obtain a target image, and transmits the target image to the display device; The display device receives the target image and displays it.

15. The image processing system according to claim 14, wherein: The display device is a projection screen or a display terminal having a resolution consistent with that of the target image.

Images